arXiv subjects

Guangyu Wei

Publications and source records attributed to Guangyu Wei.

11 recordsLinked to original sources

arXiv2026

Uncertainty-Aware Concept and Motion Segmentation for Semi-Supervised Angiography Videos

Segmentation of the main coronary artery from X-ray coronary angiography (XCA) sequences is crucial for the diagnosis of coronary artery diseases. However, this task is challenging due to issues such as blurred boundaries, inconsistent radiation contrast, complex motion patterns, and a lack of annotated images for training. Although Semi-Supervised Learning (SSL) can alleviate the annotation burden, conventional methods struggle with complicated temporal dynamics and unreliable uncertainty quantification. To address these challenges, we propose SAM3-based Teacher-student framework with Motion-Aware consistency and Progressive Confidence Regularization (SMART), a semi-supervised vessel segmentation approach for X-ray angiography videos. First, our method utilizes SAM3's unique promptable concept segmentation design and innovates a SAM3-based teacher-student framework to maximize the performance potential of both the teacher and the student. Second, we enhance segmentation by integrating the vessel mask warping technique and motion consistency loss to model complex vessel dynamics. To address the issue of unreliable teacher predictions caused by blurred boundaries and minimal contrast, we further propose a progressive confidence-aware consistency regularization to mitigate the risk of unreliable outputs. Extensive experiments on three datasets of XCA sequences from different institutions demonstrate that SMART achieves state-of-the-art performance while requiring significantly fewer annotations, making it particularly valuable for real-world clinical applications where labeled data is scarce. Our code is available at: https://github.com/qimingfan10/SMART.

cs.CV↗

arXiv2025

Towards Real-Time Fake News Detection under Evidence Scarcity

Fake news detection becomes particularly challenging in real-time scenarios, where emerging events often lack sufficient supporting evidence. Existing approaches often rely heavily on external evidence and therefore struggle to generalize under evidence scarcity. To address this issue, we propose Evaluation-Aware Selection of Experts (EASE), a novel framework for real-time fake news detection that dynamically adapts its decision-making process according to the assessed sufficiency of available evidence. EASE introduces a sequential evaluation mechanism comprising three independent perspectives: (1) Evidence-based evaluation, which assesses evidence and incorporates it into decision-making only when the evidence is sufficiently supportive; (2) Reasoning-based evaluation, which leverages the world knowledge of large language models (LLMs) and applies them only when their reliability is adequately established; and (3) Sentiment-based fallback, which integrates sentiment cues when neither evidence nor reasoning is reliable. To enhance the accuracy of evaluation processes, EASE employs instruction tuning with pseudo labels to guide each evaluator in justifying its perspective-specific knowledge through interpretable reasoning. Furthermore, the expert modules integrate the evaluators' justified assessments with the news content to enable evaluation-aware decision-making, thereby enhancing overall detection accuracy. Moreover, we introduce RealTimeNews-25, a new benchmark comprising recent news for evaluating model generalization on emerging news with limited evidence. Extensive experiments demonstrate that EASE not only achieves state-of-the-art performance across multiple benchmarks, but also significantly improves generalization to real-time news. The code and dataset are available: https://github.com/wgyhhhh/EASE.

cs.CL↗

arXiv2025

Excitation of toroidal Alfv\'en eigenmode by energetic particles in DTT and effect of negative triangularity

A linear gyrokinetic eigenvalue code is developed to study the stability of toroidal Alfv\'en eigenmode (TAE) in general axisymmetric toroidal geometry, with the self-consistent treatment of energetic particle drive and core plasma Landau damping in a non-perturbative way. The general particle responses of both circulating and trapped particles are incorporated in the calculation by means of the action-angle approach, and, particularly, the finite Larmor radius and orbit width effects of energetic particles are fully taken into account. The ballooning-mode representation is adopted to solve the eigenmode equations in order to reduce the computational resource while obtaining a high resolution of the fine radial structure. Furthermore, the code is able to study the physics of wave-particle interaction in great detail, thanks to the development of systematic theory-based numerical diagnostics, including effective mode structure and phase space resonance structure. As an application of the code, we perform an in-depth study of the triangularity effect on TAE stability based on the reference equilibrium of the Divertor Tokamak Test facility. It is demonstrated that TAE growth rate can be affected by the triangularity through the modifications of geometric couplings, resonance condition, as well as mode frequency and mode structure. As a result, negative triangularity can either stabilize or destabilize the energetic particle driven TAE depending on the dominant mechanism. The relative importance of these mechanisms under different circumstances is systematically analyzed, providing clear physical insights. The overall effect of negative triangularity for a specific tokamak scenario can be assessed based on these studies.

physics.plasm-ph↗

arXiv2025

Bi-Grid Reconstruction for Image Anomaly Detection

In image anomaly detection, significant advancements have been made using un- and self-supervised methods with datasets containing only normal samples. However, these approaches often struggle with fine-grained anomalies. This paper introduces \textbf{GRAD}: Bi-\textbf{G}rid \textbf{R}econstruction for Image \textbf{A}nomaly \textbf{D}etection, which employs two continuous grids to enhance anomaly detection from both normal and abnormal perspectives. In this work: 1) Grids as feature repositories that improve generalization and mitigate the Identical Shortcut (IS) issue; 2) An abnormal feature grid that refines normal feature boundaries, boosting detection of fine-grained defects; 3) The Feature Block Paste (FBP) module, which synthesizes various anomalies at the feature level for quick abnormal grid deployment. GRAD's robust representation capabilities also allow it to handle multiple classes with a single model. Evaluations on datasets like MVTecAD, VisA, and GoodsAD show significant performance improvements in fine-grained anomaly detection. GRAD excels in overall accuracy and in discerning subtle differences, demonstrating its superiority over existing methods.

cs.CV↗

arXiv2024

Indirect nonlinear interaction between toroidal Alfvén eigenmode and ion temperature gradient mode mediated by zonal structures

The indirect nonlinear interactions between toroidal Alfvén eigenmode (TAE) and ion temperature gradient mode (ITG) are investigated using nonlinear gyrokinetic theory and ballooning mode formalism. More specifically, the local nonlinear ITG mode equation is derived adopting the fluid-ion approximation, with the contributions of zonal field structure and phase space zonal structure beat-driven by finite amplitude TAE accounted for on the same footing. The obtained nonlinear ITG mode equation is solved both analytically and numerically, and it is found that, the zonal structure beat-driven by TAE has only weakly destabilizing effects on ITG, contrary to usual speculations and existing numerical results.

physics.plasm-ph↗

arXiv2024

Effects of plasma nonuniformity on zero frequency zonal structure generation by drift Alfven wave instabilities in toroidal plasmas

Effects of plasma nonuniformity on zero frequency zonal structure (ZFZS) excitation by drift Alfven wave (DAW) instabilities in toroidal plasmas are investigated using nonlinear gyrokinetic theory. The governing equations describing nonlinear interactions among ZFZS and DAWs are derived, with the contribution of DAWs self-beating and radial modulation accounted for on the same footing. The obtained equations are then used to derive the nonlinear dispersion relation, which is then applied to investigate ZFZS generation in several scenarios. In particular, it is found that, the condition for zonal flow excitation by kinetic ballooning mode (KBM) could be sensitive to plasma parameters, and more detailed investigation is needed to understand KBM nonlinear saturation, crucial for bulk plasma transport in future reactors.

physics.plasm-ph↗

arXiv2024

Nonlinear saturation of reversed shear Alfven eigenmode via high-frequency quasi-mode generation

A nonlinear saturation mechanism for reversed shear Alfven eigenmode (RSAE) is proposed and analysed, and is shown to be of relevance to typical reactor parameter region. The saturation is achieved through the generation of high-frequency quasi-mode due to nonlinear coupling of two RSAEs, which is then damped due to coupling with the shear Alfven continuum, and leads to the nonlinear saturation of the primary RSAEs . An estimation of the nonlinear damping rate is also provided.

physics.plasm-ph↗

arXiv2024

A Deep Learning Model for Coronary Artery Segmentation and Quantitative Stenosis Detection in Angiographic Images

Coronary artery disease (CAD) is a leading cause of cardiovascular-related mortality, and accurate stenosis detection is crucial for effective clinical decision-making. Coronary angiography remains the gold standard for diagnosing CAD, but manual analysis of angiograms is prone to errors and subjectivity. This study aims to develop a deep learning-based approach for the automatic segmentation of coronary arteries from angiographic images and the quantitative detection of stenosis, thereby improving the accuracy and efficiency of CAD diagnosis. We propose a novel deep learning-based method for the automatic segmentation of coronary arteries in angiographic images, coupled with a dynamic cohort method for stenosis detection. The segmentation model combines the MedSAM and VM-UNet architectures to achieve high-performance results. After segmentation, the vascular centerline is extracted, vessel diameter is computed, and the degree of stenosis is measured with high precision, enabling accurate identification of arterial stenosis. On the mixed dataset (including the ARCADE, DCA1, and GH datasets), the model achieved an average IoU of 0.6308, with sensitivity and specificity of 0.9772 and 0.9903, respectively. On the ARCADE dataset, the average IoU was 0.6303, with sensitivity of 0.9832 and specificity of 0.9933. Additionally, the stenosis detection algorithm achieved a true positive rate (TPR) of 0.5867 and a positive predictive value (PPV) of 0.5911, demonstrating the effectiveness of our model in analyzing coronary angiography images. SAM-VMNet offers a promising tool for the automated segmentation and detection of coronary artery stenosis. The model's high accuracy and robustness provide significant clinical value for the early diagnosis and treatment planning of CAD. The code and examples are available at https://github.com/qimingfan10/SAM-VMNet.

eess.IV↗

arXiv2024

Calculation of toroidal Alfvén eigenmode mode structure in general axisymmetric toroidal geometry

A workflow is developed based on the ideal MHD model to investigate the linear physics of various Alfvén eigenmodes in general axisymmetric toroidal geometry, by solving the coupled shear Alfvén wave (SAW) and ion sound wave (ISW) equations in ballooning space. The model equations are solved by the FALCON code in the singular layer, and the corresponding solutions are then taken as the boundary conditions for calculating parallel mode structures in the whole ballooning space. As an application of the code, the frequencies and mode structures of toroidal Alfvén eigenmode (TAE) are calculated in the reference equilibria of the Divertor Tokamak Test facility (DTT) with positive and negative triangularities, respectively. By properly handling the boundary conditions, we demonstrate finite TAE damping due to coupling with the local acoustic continuum, and find that the damping rate is small for typical plasma parameters.

physics.plasm-ph↗

arXiv2022

The enhanced soliton propagation and energy transfer in the coupled drift wave and energetic-particle-induced geodesic acoustic mode system

The evolution of the coupled drift wave (DW) and energetic-particle-induced geodesic acoustic mode (EGAM) nonlinear system is investigated using the fully nonlinear coupled DW-EGAM two-field equations, with emphasis on the turbulence spreading in the form of soliton and the nonlinear energy transfer between DW and EGAM. Four scenarios with different combinations of EGAM initial amplitudes and linear EGAM growth rates are designed to delineate the effects of linear EGAM drive and finite EGAM amplitude on DW nonlinear dynamic evolution. In presence of the linear EPs drive, the soliton propagation is enhanced, due to the generation of small radial scale structures. Two conservation laws of the nonlinear system are derived, including the energy conservation law. It is found that the energy of DW always decreases and that of EGAM always increases, leading to regulation of DW by EGAM.

physics.plasm-ph↗

arXiv2021

Soliton generation and drift wave turbulence spreading via geodesic acoustic mode excitation

The two-field equations governing fully nonlinear dynamics of the drift wave (DW) and geodesic acoustic mode (GAM) in the toroidal geometry are derived in nonlinear gyrokinetic framework. Two stages with distinctive features are identified and analyzed. In the linear growth stage, the set of nonlinear equations can be reduced to the intensively studied parametric decay instability (PDI), accounting for the spontaneous resonant excitation of GAM by DW. The main results of previous works on spontaneous GAM excitation, e.g., the much enhanced GAM group velocity and the nonlinear growth rate of GAM, are reproduced from numerical solution of the two-field equations. In the fully nonlinear stage, soliton structures are observed to form due to the balancing of the self-trapping effect by the spontaneously excited GAM and kinetic dispersiveness of DW. The soliton structures enhance turbulence spreading from DW linearly unstable to stable region, exhibiting convective propagation instead of typical linear dispersive process, and is thus, expected to induce core-edge interaction and nonlocal transport.

physics.plasm-ph↗